ClickHouse: High-Performance Distributed DBMS for Analytics
Yandex team has built one of the best opensource databases for analytics. It's fast, capable for storing petabytes of data and supports SQL. This talk is ClickHouse overview: features and benchmarks, plans and statuses, use cases and real users feedback. I will start with current ClickHouse status in DBMS market: how many users, what's the community size, and what's going on in contributing. In the main part I will cover up key database features and capabilities. Some benchmarks will be shown with most interesting competitors and I'll tell how anybody can reproduce results to verify and add competitors. The main question we have from everybody: "Why it's so fast? It can't be so fast!". I'll explain main design ideas and will answer this question. Then I will show most profitable use cases for both big and small companies. We already have feedback from external companies and have cases of great success. Success in building new products; in saving money; in opening new ways of analysing business data. At the end I will share our current development pipeline and future plans.
Head of ClickHouse development team, ClickHouse
Head of Analytics Systems Department, Yandex
Hardware background: HDL and microprocessors. Work in Yandex since 2012 as developer, in Yandex.Metrica team. New backend architecture design, a lot of code written. Responsible for analytic products development since September 2015.